Home  |  Search  |  Contents  |  Introduction  |  Login
Evolution of the Edition | Textual Introduction | Principles of Selection | Works Cited

But, whether a poem is approached through the eyes in a book, or through the ears, the eyes within the eyes, the visual imagination, are reached; and this in itself is a way of reaching the total imagination.

—Muriel Rukeyser, The Life of Poetry

This introduction is a biography of Emily Dickinson’s Correspondences: A Born-Digital Textual Inquiry (hereafter cited as EDC). Thus it recounts EDC’s evolution from its inception as a digital archive of all of Dickinson’s writings to a born-digital critical edition of selected documents, designed to facilitate the kinds of interaction between readers and scholarly texts in ways that Smith’s first book, Rowing in Eden: Rereading Emily Dickinson, only dreamed. (Those dreams are laid out in the book’s second chapter, “Rowing in Eden: Reading Dickinson Reading,” available online at http://www.emilydickinson.org/resources/smith_rowing/rowing1.html.) Of this introduction, Smith is the lead author of section 1, Evolution from “Comprehensive” Archive to Selected Critical Edition, and section 3, A Brief Note on the Principles of Selection, which inform the decision to choose these particular thirty Emily Dickinson texts, texts that were first published bibliographically in or before 1924. (Also included are all of their known original writings as well as, in an update scheduled for the spring, the known printings.) Lara Vetter is the lead author of section 2, an essay on principles of editing and markup, “Dickinson’s Letters in “broken mathematics”: A Textual Introduction.” As is the case for the editorial notes throughout this edition, Ellen Louise Hart serves as contributing consultant to this introduction.

I. Evolution from “Comprehensive” Archive to Selected Critical Edition

EDC is the major editorial work on Emily Dickinson’s writings being produced by the Dickinson Electronic Archives (DEA) [projects and EDC’s evolutions necessarily include those of the DEA]. Launched in 1995, the DEA (importantly, not the “Emily Dickinson Electronic Archives”) is an online resource of hypertextual editions of writing by Dickinson as well as by her family, of contemporary poets and scholars responding to her legacy, and of pedagogical sites helpful for teaching Dickinson, Whitman, and nineteenth-century American literature and culture. EDC is an online scholarly edition of selected writings by Dickinson, marked up in TEI-conformant XML and published by the University of Virginia Press. In effect, though developed before the “Database as Genre” forum appeared in PMLA in fall 2007, our design for EDC heeds the superb advice of scholars such as Peter Stallybrass, whose wise reminder in that exchange is to “spend more time on less” (1584); of Jerome McGann, who in the same feature notes that database and anything marked up for the digital world “is severely constrained and organized” (1588); and of Meredith McGill, who points out that the “promise of completeness” is simply that, a promise for and not a fact of the matter, and concludes, “If we misconstrue media shift as liberation, we are likely to settle for less than the new technologies can offer us” (1595). The rapid advancements in the development of software and hardware for use in humanities scholarship have made EDC and new editorial work, work formerly constrained by the printed book, possible.

Yet the most important technology employed in this edition is not the highly structured markup used for effective storage and retrieval, or the high-quality digital images of Dickinson’s manuscripts, or the Zoomify and other software that readers can use to examine both the details of documents and their relationships to one another, but what Smith has elsewhere been actively promoting: the “technology of self-consciousness.” Encoding of texts for computationally-driven digital distribution requires a technology of self-consciousness that can in turn produce a healthy self-consciousness about what Bruno Latour and Steve Woolgar describe in Laboratory Life as “black-boxing”—which occurs when one “renders items of knowledge distinct from the circumstances of their creation” (259n; see also, Smith, “Computing: What’s American Literature Study Got to Do with IT?”). In black-boxing, critical opinion becomes “fact”; more often than not, amnesia sets in after that factual instantiation, and having been effectively black-boxed, “fact” becomes “truth.” Bibliographic editions are particularly susceptible to black-boxing because many decisions made are not described by any of the notes accompanying such productions, and though digital editions have an opportunity to make visible decision-making processes that bibliographic editions cannot, they do not necessarily do so, and many, “hewing surprisingly closely to normative ideas of the author and the work,” retain a “conceptual and structural horizon that keeps such projects from functioning” in editorially radical ways (McGill 1593). So digital editions tend not to avail themselves of this unprecedented opportunity for facilitating transparency about production and involving users in editorial processes, as EDC aims to do. Rather than “thicken[ing] the medium as much as possible” by “literally . . . put[ting] the resources of the medium on full display” in order “to exhibit the processes of self-reflection and self-generation which texts set in motion” (McGann, Textual Condition 14), the seductive power of scope—the vast number of texts that can be featured online, for example—has too often overshadowed advancing methods of critical inquiry. Maintaining relentless self-consciousness about how critical “facts” have been produced, about how items of knowledge are part of the circumstances of their creation, is crucial for responsibly providing the provisionality that characterizes the best editorial science. Even after the extensive work in digital humanities in the 1990s and early 2000s, much about our foundational materials—who made the texts you see on the screen before you and for what purposes— is still taken for granted, so much so that in December 2007 an esteemed Dickinson scholar unself-consciously declared from the stage of the Folger Shakespeare Library an edition of her writings the “definitive standard.” An acute self-consciousness about faith in such definitude is vital for providing democratizing access to primary materials while maintaining intellectual rigor. In producing EDC, one of our primary goals has been to thicken this editorial medium, and we welcome suggestions from readers about more ways in which to achieve this objective.

More than fifteen years ago, before Harvard University Press produced a new variorum of Dickinson’s poems, before Hart and Smith made the book edition Open Me Carefully: Emily Dickinson’s Intimate Letters to Susan Huntington Dickinson, and before we were aware that there could be anything called an electronic archive, Smith noted that “neither the reproductions of texts nor critical interpretations can be innocent of or superior to politics, since both require negotiations among authors, editors, publishers, and readers. Dickinson interpretation will be powerfully enhanced by cultivating constant awareness of the ‘official’ repatternings of the variorum, the three-volume letters, and the separate publication of the ‘Master’ documents” (Smith, Rowing in Eden 94; see also Rowing, ‘Book’ Making: Reading is Always a Matter of (Re)Patterning”). Our work on the electronic archives projects has only reinforced these words (as it has provided us with tools to make such critical investigations), and the exhortative words about cultivating constant awareness of editorial repatternings, micro and macro alike, apply to the evaluation of all editions, including the one on the screen before you.

The DEA projects that generated EDC grew out of Rowing in Eden, which was printed in 1992. Almost immediately upon its publication, Smith fell into serious conversation with Jerome McGann about the importance of producing what we were then calling a “hypermedia archive” of digital surrogates of all of Dickinson’s manuscripts, which are scattered among several different repositories. If they were gathered together via hyperlinking among the images and deeply encoded (now via TEI-conformant XML; then SGML) text, and made available via electronic distribution, release of the kind of imaginative power that had come from Smith’s own critical/archival explorations could be made available to far more scholars, indeed to any reader with access to a personal computer and the World Wide Web. Through this kind of distribution providing views heretofore impossible to the reader without access to the physical archives, readers would have much more extensive exposure to the range and the diversity of Dickinson’s writings, which have been conflated and homogenized in previous editions as “poems” produced in two different variorum editions and in a two-volume facsimile edition of her manuscript books, and as “letters” in a three-volume edition (in Life and Letters of Emily Dickinson, her niece Martha Dickinson Bianchi did print some writings as letter-poems, but her practice and that found in Hart and Smith’s Open Me Carefully are the exceptions to the rule sharply separating poetry from prose).

As those scholarly editions claim to be, the Dickinson Electronic Archives team first imagined a digital scholarly edition that would be “comprehensive.” The plan was to edit every known writing of Emily Dickinson for digital distribution and, for our organizational strategy, to follow her method of coterie publication—in which she distributed poems in letters to at least a hundred correspondents and bound them into manuscript books found after her death, presumably for “Posterity.” In other words, instead of separating her writings into “poems” on the one hand and “letters” on the other, we planned to present them in simulacra of the containers in which she left them. Those “containers” are not the writings themselves but are the means by which Dickinson distributed those texts and documents or left them behind—via envelope through the postal service, via personal carrier with stationery folded for simple hand delivery, via handmade books and on sheets with other writings, via handmade books and on a leaf gathered with other writings, on a half sheet or leaf torn from a sheet that folded makes four pages, torn in half makes two (for further clarification about texts themselves not being containers, see McGann, Radiant Textuality 2; in the case of Dickinson, documents reveal the “containers” of transmission though the texts themselves are not containers). The other “containers” we wished to remind readers of are the “constellations”, variants of texts featured in different fora; see our textual introduction. Our plan was to begin with her most voluminous correspondence to Susan Huntington Gilbert Dickinson, then turn to her writings to her next-most-frequently-addressed audience, Thomas Wentworth Higginson, and proceed to work through, correspondence-by-correspondence, in order of the number of documents sent to an audience, so that the least voluminous correspondence would be edited last. Yet issues with copyright and claims of textual ownership made realizing the dream of that plan problematic. Also, even before those copyright and permissions issues mandated that the plan for this Rotunda edition be redrawn, the Dickinson Editing Collective (the archive development team) realized some important facts about archives that would have to be self-consciously highlighted in any digital scholarly edition; and so we did not see the Harvard University Press copyright and permissions issues only as thwarting our goals but as opportunities to think much harder about archival issues rather than simply being swept away by the promise of a seemingly unbound medium.

The DEA are virtual and house-digital surrogates of manuscripts as well as born-digital articles and bibliographic articles electronically distributed. As do physical archives in libraries, the Dickinson Electronic Archives are epistemic arrangements that do not house what was—in this case writings by Emily Dickinson—but rather some of what remains of her writings, some of what has survived and is now housed in the Houghton at Harvard, the Frost at Amherst College, the Jones Library in Amherst, the Library of Congress, the Rosenbach in Philadelphia, the Morgan in New York, and various other repositories. Immensely exciting about the original goal of the Dickinson Electronic Archives scholarly edition was to collect—by deep linking and markup—this diaspora of surviving Dickinson documents, gesturing toward the promise of completeness that McGill rightly remarks as problematic. That promise temporarily obscured what are in fact vitally important critical achievements that can be facilitated in the digital realm, achievements that do not depend on the romance of comprehensivity. That mind-blowing capacity for gathering together that which had been scattered can distract one from posing questions about the archival logics of the physical and virtual archives and about the archival practices both informed by and informing those logics. As the earliest articles of Smith and others about digital archives make plain, a sense of frenzied finding similar to that which often fuels the excitement of a neophyte researcher characterized the early promises and hopes of many archives builders—look at what can be seen, look at what we can gather together—and indeed still characterizes the claims of some digital enthusiasts. For example, as several of the critiques in the aforementioned PMLA exchange note, optimistic claims for the democratizing power of databases elide the occlusions of taxonomizing ideologies and of the rigid categories on which databases depend, as well as the interpretive structures of metadata. Yet as the evolutions of the Dickinson Electronic Archives to include others’ writings, and critical, poetic, and pedagogical responses show, the Dickinson Editing Collective (producers of the Dickinson Electronic Archives) realized in the earliest stages of digital development that examining how we use the findings enabled by these new collections and critiquing what had been staged in previous editions, as well as what is being staged in our editions, are of utmost importance. The power of scope and representation that first threatened to overshadow them soon gave way to the questions that should be central to all editions: how did these items of knowledge come into being, who made them, for what purposes, and what can we learn from their survival? Thus, fortuitously, the proprietary claims of Harvard University Press and our own critical interests in editing, editorial theories, and editorial ideologies walked hand-in-hand, as it were, and slowed the originally conceived project.

The archives gathered by our attentions foreground an important point, one that Smith urged taking into account in discussing “Dickinson-Johnson” poems and “Dickinson-Franklin” poems (126–127) in her essay on “Dickinson’s Manuscripts” in The Emily Dickinson Handbook (1998), and one that all of our introductions and notes throughout the DEA spotlight: editors always collaborate with writers in the making of authorial texts, and editorial interests cannot help but have a profound impact on how editions are conceived and constituted. DEA projects were generated first by Smith’s critical interests (she proposed the project to the Institute of Advanced Technology in the Humanities (IATH) at the University of Virginia, has written numerous grant proposals to fund the work itself, and proposed the digital scholarly edition to Rotunda), and began as archives of her attentions, epistemic choreographies shaped by her arguments (as the 1998 variorum was shaped by the archives of Franklin’s attentions and his editorial stances). In the case of the Dickinson Electronic Archives, their design is to be more and more inclusive (convening a Collective for their input, inviting Open Critical Review and user participation). We have invited contributions from Dickinson scholars with oppositional views about best editorial practices as far as her writings are concerned, and will continue to do so. A wiki or blog will accompany EDC, the kind of interactive tool that would be more than welcomed as part of a digital scholarly edition (as is indicated by discussions on the Digital Americanist listserv, February 2008). Important foci of this Rotunda edition include questions of transparency regarding editorial projects and their biases and the critique of editorial and reading practices, including our own.

Comprehensivity and completeness are laudable goals, and we retain those objectives for the documents reported upon in this edition. Yet we do so with the awareness that such aims are impossible dreams, and for a variety of reasons. For one thing, a new Dickinson manuscript has been discovered since the printing of the variorum and additional manuscripts may well come to light, a simple fact of the matter that underscores that archives are not collections of what was but of what remains. Yet even more importantly, Dickinson’s texts have been and continue to be circulated in a variety of perpetually evolving forms, and those contesting circulations offer vital examples and data about various principles of reading, of how poetry lives in and across diverse readerships, and of the crucial (and often invisible) role editorial praxes play in shaping both reading practices and the communities and individuals who exchange ideas in them.

So as important as those goals is the ability of a digital edition to turn editorial processes inside out, to pull back the curtain, as it were, on how decisions are reached to make “Emily Dickinson poems” and “Emily Dickinson letters.” The Dickinson Electronic Archives shows the importance of embracing textual fluidities, even (or maybe especially) for readers who insist there must be one “standard” editorial transmission, for readers who believe that what was envisioned by the poet can be definitively translated into print (see Bryant’s The Fluid Text and his “Witness and Access”). The digital articles produced in the last decade of the twentieth century through the early years of the twenty-first century all attest to this principle (and the plan for digitally editing all of Dickinson’s writings would have been executed keeping faith with that). When the copyright claims of Harvard University Press inhibited the scope of that original plan, we agreed, after detailed negotiations, that a selected edition of her writings could be produced to everyone’s satisfaction. By carefully selecting thirty of her texts for digital editing, enough can be made known about the variety of her writings so that most of the types of Dickinson’s manuscripts can be presented and critically plumbed. So as a critical examination of editorial praxes, this digital edition of selected writings originating from the hand of Emily Dickinson seeks to reorient readers about what a scholarly edition can be.

For a couple of decades Smith has been arguing that triangular intertextualities—the dynamic influences of biography, reception, and textual reproduction upon one another—inform all editorial practices; indeed, they inform all critical positions and positionings. Editions produced for print effectively “black box” research methods so that the contact between the user and the producer is cloaked in mystery. This has been true as far as Dickinson editions go (including that of Hart and Smith), though the editions produced since Harvard’s first variorum in 1955 have been more and more transparent, or at least have tried to be so. Even in the early twenty-first century, however, the old truisms many critics still assume of scholarly publishing and editorial work, the ones entangled in and by the social relations of the book, still hold in Dickinson studies, as a series of articles and books that resist manuscript study and tout the more hieratic, authoritarian view that one editor’s work should be the standard make plain. The bibliographically sealed position asserted as “best” by some within and without Dickinson studies is one Smith has described before in “The Human Touch, Software of the Highest Order: Revisiting Editing as Interpretation.” These erroneous assumptions assert the following:

As far as editorial work goes, such assumptions lead to critical games of “gotcha” among editors and critics and suppress the creation and validity of different versions, which may allow that both, neither, or either might all be true. Neither faultfinding as an end in itself nor suppression by hiding the processes that determined final products is healthy for knowledge production and critical understanding.

Moving beyond the idea that scholarly conclusions can only be authentic, authoritative, and reliable when determined by a single expert or group of experts makes it possible to bring in keen insights from people with different levels of expertise. In the case of the DEA experiments, middle-school teachers and their students have offered telling insights. Inspired by the poet’s reworkings of “Safe in their Alabaster Chambers” studied in “Emily Dickinson Writing a Poem,” the first digital article produced for the DEA, seventh-grade students wrote poems of their own that extended a metaphor just as Dickinson herself did in ‘Hope’ is the thing with feathers” (FP 314) and contributed them to the “Contemporary Youth’s Companion” section of the Web site. That is one serendipitous result of the Dickinson Electronic Archives early experiments: having realized that poems do not spring fully formed from a poet’s mind, seventh graders were inspired not only to read but also to write poetry. Their activity parallels the “writing” that all readers and all editors do in the acts of reading and of textual production, the very sort of readerly writing that conventional editions, however scrupulously produced, occlude. In this dynamic, perpetually updatable environment, our ambitions for the EDC is that our readers engage in conversations about editorial praxes rather than stake out seemingly intractable positions of “better” or “right” or “wrong.”

So what new methods of editing and understanding texts might, with diligent application of human software, be created in a dynamic electronic environment such as this created for Emily Dickinson’s Correspondences: A Born-Digital Textual Inquiry? Once again, we draw on observations made previously by Smith, most recently in “The Human Touch, or Software of the Highest Order”:

Thus this scholarly digital publication makes no claim to be “definitive,” nor does it aim to supplant any other edition, including those produced by Harvard University Press. This digital publication is in conversation with R. W. Franklin’s and Thomas Johnson’s editions, as well as with the Amherst microfilms, and editions by Martha Dickinson Bianchi, Mabel Loomis Todd and Millicent Todd Bingham, as well as by Richard Sewall [see the Blackwell Companion Bibliography], Werner, and Hart and Smith. Each of the thirty texts selected for this edition was sent to Susan Dickinson, Emily Dickinson’s primary correspondent and the only contemporary reader for whom Dickinson altered a poem. Though Hart and Smith’s focus for the past fifteen years has been on that primary correspondence with Susan Dickinson, and we use that corpus in order to show the diversity of Emily Dickinson’s writings (which, unlike Dickinson’s other correspondences, feature all of the various kinds of writings we know Dickinson generated, including the material differences found in scraps), comparison to what was sent to other audiences is necessary for formulating profoundly probative analysis, especially if one is to achieve well-rounded critical descriptions and interpretations. So we have selected thirty core texts sent to Susan Dickinson to show the diversity of documents sent to her and featured in the Dickinson corpus. We also present all known versions of those core texts, and each is embedded within constellations of alternate versions and related documents. While this volume may appear lesser in scope than the one originally conceived, Emily Dickinson’s Correspondences: A Born-Digital Textual Inquiry (EDC) delves more deeply into the questions that should be asked of and by any edition: who made the texts on the screen or page before you and for what purposes?

II. Dickinson’s Letters in “broken mathematics”: A Textual Introduction

In broken mathematics

We Estimate our prize

Vast - in it’s fading ratio

To our penurious Eyes! (H 2, FP 78)

The editorial achievements printing the manuscript writings of Emily Dickinson have been most impressive. Since the first posthumous volume was published near Christmas 1890, twenty-nine more or less official volumes of Dickinson’s works have appeared, printing the writings in EDC in a variety of forms. The DEA projects build on this superb foundation by presenting our critical inquiries regarding the editing of Emily Dickinson’s writings via new media and for open critical review, by incorporating new media tools into scholarly editing and thus using previously unavailable strategies for representation of original materials and for storage and retrieval of the consequent data. Moreover, this volume differs from previous ones in the attention paid to representing Dickinson’s poems in their original contexts (often in letters sent to her many correspondents) and to privileging as much the material page as the conceptual one.

With the exception of Martha Dickinson Bianchi’s Life and Letters of Emily Dickinson, which presented some letter-poems, until Open Me Carefully, printed volumes of Dickinson have represented either poems or letters, not both, therefore organizing the materials by dividing prose from poetry. Virginia Jackson’s immensely proliferative, incisive query regarding Dickinson’s work and the editorial makings of her poems is the latest in a series of scholarly critiques that make clear the pitfalls of such an approach: “Can a text not intended as a lyric become one? Can a text once read as a lyric be unread? If so, then what is—or was—a lyric?” (6). By contrast, the Dickinson Editing Collective’s organizing principle is guided by the record left by Dickinson’s writing and distribution practices (individual manuscripts sent to a particular audience, and saved and stored by that recipient; writings saved and stored by Dickinson herself—either bound into a manuscript book and tied with a string or left loose in a box or drawer). The work of the Collective is also set apart by the fact that our editorial center is Dickinson’s manuscript page and/or sheet, and so images of those pages/sheets are key elements of our edited reproductions. With editors such as Thomas H. Johnson and R.W. Franklin, the Collective shares a strong commitment to producing resources of the highest achievable quality. To this end, we seek to render representations that faithfully report what is seen on the page. Additionally, we aim to transmit Dickinson’s writing in ways that self-consciously exploit interpretive noise as much as possible in order to make visible the messiness of editing and to reveal the fallacy of the “pure” text. The primary work of EDC is to make data about the poet’s writing and epistolary distribution as available and usable as is currently possible.

Detailing the goals of EDC is probably best accomplished by describing the contrasts between EDC and an intimately related but distinctly different editorial endeavor, a print production of some of the materials from Dickinson’s primary correspondence, Open Me Carefully: Emily Dickinson’s Intimate Letters to Susan Huntington Dickinson. In Open Me Carefully the goal of editors Smith and Hart was to make “a cohesive book that would most effectively relate the human story behind this most generative of literary and emotional unions” between Susan and Emily Dickinson. Painfully aware that “typography cannot sufficiently transmit . . . aspects of Dickinson’s handwriting which influence[d]” their editing, Hart and Smith translated line breaks and some marks of punctuation into print and noted as much as they could in the notes at the back of the book (given the publishing limitations they faced) but did not pretend to present a representation of many of the material elements of manuscript inscriptions such as oddly shaped punctuation marks, alphabetic letters that appear to be middle rather than upper or lower case, alphabetic letters whose shape and ligation make one wonder, as Franklin did, if Dickinson was writing “opon” rather than “open” or “upon” (OMC xxii–xxiv). Theirs was a selection to tell the story of a richly productive literary liaison and was presented as such, and Hart and Smith realized that to attempt to translate such matters into print would be cumbersome and distract from the primary goal of that volume.

EDC does not have the same interpretive drive as did Open Me Carefully. By contrast, EDC’s interpretive drive is to use the simulacrum of Dickinson’s placement of her writings in the aforementioned containers—poems blended into or embedded into the prose in letters; poems written on separate sheets and enclosed with letters; letters with no poems enclosed or embedded; letters that are also poems—and of their placement in constellations to tell stories about how Emily Dickinson poems have been made and unmade. By using her writings as an example, EDC examines the powers and perils of editing, and tests ways in which editing with digital tools and in the digital realm might make editorial praxes more transparent and more available to analysis. Part of that transparency is to make our editorial choices as clear as possible, including choices in markup. The markup for each text is made available (in the lower left-hand corner of every document’s presentation), and all data in this edition will be continually reviewed and updated.

Representing Emily Dickinson’s manuscripts in a scholarly electronic edition requires both the formulation of an editorial statement—outlining the principles of editing the manuscripts—and an additional layer of translation not required of editors of print editions, who do have to consider issues of display but not underlying structure and relationships between parts of a piece of writing and the document on which it is inscribed. This additional step involves transmuting those principles of editing into an encoding scheme compatible with the guidelines of the Text Encoding Initiative (TEI), the institutional body that creates and maintains standards of literary editing in electronic environments. For years, Smith has written persuasively about the need for an electronic edition of Dickinson’s manuscripts, especially her correspondence; to date, only Radical Scatters, Werner’s edition of some of Dickinson’s fragments, has attempted such an endeavor. (Werner is currently revising Radical Scatters for publication through the University of Nebraska’s Center for Digital Research in the Humanities and through NINES, through which EDC also plans to work, as has the Rotunda edition of Melville’s Typee, edited by John Bryant). Smith and Werner’s works in theorizing Dickinson in digital contexts have served as important models and inspirations for EDC.

While Smith, Hart, Werner, and Vetter discussed editorial challenges and frameworks for years prior to compiling this edition—deciding ultimately upon a set of principles that represent at once staunch agreements, passionate disagreements, and careful compromises—the work of the edition has generally followed a division of labor: Smith overseeing the editorial work, and Vetter overseeing the encoding (or “markup”) that rests invisibly beneath the text that appears on the screen. However, despite this division it is crucial to note that our work overlaps in significant ways. Editorial decisions concern structural and ontological categories of literary texts—what is a line? what is a poem?—and encoding is concerned with precisely these categories. Likewise, encoding decisions are complexly implicated in theories of editing; as one of the TEI’s creators, Michael Sperberg-McQueen, observes, “Markup reflects a theory of text” (35). Electronic editions are no different, as Sperberg-McQueen contends: “What computers process are representations of data. . . . Representations are inevitably partial, never disinterested” (34). Editions are not neutral objects, but the process of scholarly editing is rarely made transparent. Readers, we believe, should be made aware of the editorial interventions made in the process of producing an edited text; thus, “technical” issues are discussed below in terms of their relationship to editorial theory as a way of both educating those with little expertise in digital studies and uncovering the layer of markup that underlies the texts. The inevitable overlap between the two introductions that are now fused together into this iteration nicely mirrors the intersecting spheres of work on this edition.

Scholarly editions, according to one recent scholar, “are more often than not producing new rhetorical documents, where both the editor her/himself and the readers of her/his edition are perfectly aware of the ways the edition text deviates from the edited document texts, and where this very reconfiguring, repositioning and recontextualizing of the edited work in fact is conceived of as a core value of the editorial work” (Dahlström 26). There are significant advantages to creating an electronic edition of Emily Dickinson’s work: full-color facsimiles of manuscript pages, unprecedented sorting and full-text searching, hypertextual linking of materials within webs of information, exposure to relatively inaccessible materials. But, as Vetter has argued elsewhere, there are also real shortcomings in representing a work of creativity and imagination within a metalanguage that, because it needs to be read and interpreted by a computer, relies on strict and rigid categories of hierarchy and structure. The challenges we faced in translating literature—“our prize / Vast”—into the “broken mathematics” of the machine, and the sometimes uneasy compromises we made, are disclosed below as well (H 2; FP 78).

Textuality, Materiality: General Principles

This edition was born in a desire to focus as much attention as possible on the material page of Dickinson’s literary productions. Situating the material page at the center of our theory of Dickinson’s text, this edition offers (reasonably) diplomatic transcriptions, retaining and rendering Dickinson’s page and line breaks, revision marks, and variants. Doing so, we record an unprecedented amount of bibliographic data about the manuscript itself—treating the manuscript as a physical artifact—including, for instance, a physical description of the paper, medium, document condition, handwriting by others, signs of handling (folds, stains), and traces of enclosures and accompanying materials, as well as information regarding the manuscript’s current location. Rather than conceiving of Dickinson’s writings in two distinct genres (poems, letters), we restore Dickinson’s poems to their contexts, placed in fascicles, on sheets unbound, and in the letters in which they were sent, either by her or, as in the case of “These are the days when birds come back,” by an editor to a reader who wanted a souvenir. In most print editions, Dickinson’s poems have been stripped from their contexts—the letters or manuscript book or sheet in and on which they initially appear—and printed in a manner that sets verse text clearly apart from the prose or physical condition that surrounds, appends, prefaces, or otherwise frames it (see Alexandra Socarides’ “The Poetics of Interruption” for analysis of the framing of texts on sheets as well as pages). By providing digital surrogates as well as detailed descriptions, we stage a presentation of her writings that gives readers a clearer sense of how Dickinson herself, as well as subsequent editors, situated and handled them. Though we retain distinctions between verse and prose writing at a layer beneath the display (liberally defined, in order to facilitate searching), our transcriptions do not signal genre shifts, and readers will have the manuscript facsimiles at hand to see the ways in which Dickinson’s prose and verse so often flow into one another, disturbing and confounding notions of clear generic boundaries. However, this focus on materiality does not mean that we have ignored conceptual units and structures: we group together pages and separate sheets that represent a single document.

In Textualterity, Joseph Grigely suggests that we might re-envision the practice of editing as one that studies difference, to “look at what occurs in the transition between texts”: “Toward this end an ideal edition might not be an edition at all, but a guide to historically situated texts, a Baedeker of the diachronic publication history of individual works” (48–49). McGann links this goal to electronic editions: “Unlike a traditional edition, a HyperText is not organized to focus attention on one particular text or set of texts. It is ordered to disperse attention as broadly as possible” (“Rationale”). Our focus on the material page means that we view multiple instantiations of the same poem as separate texts, linked as part of a constellation but autonomous. By encoding each manuscript separately and constructing constellations of linking between them, we acknowledge that each version possesses its own context and genealogy while bearing some significant relationship to other manuscripts; this practice also resists the notion that any one version is privileged over another. In the constellations, each version is linked to every other version and to each printing of the version, creating a web of related documents. (And user input may recognize webs of affiliations we have not identified.) For this set of documents presented in EDC, readers should keep in mind that the appearance of a version in her most voluminous correspondence with Susan Dickinson serves as the starting point for each of the thirty texts selected.

“Let us divide - with / Skill -” (A 92–10, FP 913): Textual Classification

Principles of inclusion in EDC are rather straightforward. In Johnson’s 1958 three-volume edition of the letters, he observes, “The noteworthy characteristic of the Dickinson letters, like that of the poems, is acute sensitivity. Indeed, early in the 1860s, when Emily Dickinson seems to have first gained assurance of her destiny as a poet, the letters both in style and rhythm begin to take on qualities that are so nearly the quality of her poems as on occasion to leave the reader in doubt where the letter leaves off and the poem begins” (Johnson xv). This admiring but somewhat exasperated remark is often repeated, for Dickinson’s commingling of genres, her epistolary poetics, is widely recognized. Realizing the need for some useful classification scheme—to make the writings searchable—and reminding ourselves that the manuscript facsimiles would be visible for users to examine for themselves, we at last decided upon principles of generic classification in EDC that are, as are our principles of inclusion, fairly straightforward:

Tags denoting structural divisions are not visible to the user but are powerful aids to searching the corpus. In the TEI, principle divisions of text are tagged as textual divisions, or <div>s. In EDC, the main division <div0> represents the manuscript document as a whole; the <div0> is typed according to one of the four genres described above. Further divisions within the <div0> will be used in the case of one genre, “letter with attached verse,” in instances where structural anomalies make it necessary (such as instances of “postscripts” or stand-alone extralinear marginalia), and in cases of multiple documents appearing on single pages. A subdivision of a <div0> is tagged with a <div1>. If the text is stand-alone extralinear marginalia, this is noted on the type attribute of the tag of the <div1>.

In poetry, the line group <lg> (typically, a stanza) denotes a line or group of lines that is separated from the rest by vertical spacing. The line tag <l> is used to record physical line breaks, rather than conceptual ones. Should editors deem it productive to tag metrical lines in the future, the metrical line tag <ml/> was created for this purpose. In prose, the <p> is used to denote a line or group of lines that is separated from the rest by vertical spacing or some other significant physical indicator. Counter to TEI usage, we use <p> to denote prose chunks, and not necessarily paragraphs in the traditional sense, since Dickinson’s prose does not always employ what we might think of as conventional paragraphs.

TEI includes tags for many parts of a letter. If present, prefatory information—such as places, dates, and greetings—is tagged within an <opener> that includes <dateline> and/or <salute> . The tag <closer> marks up closing text that may appear, including a closing salutation <salute> and signature <signed> . The TEI does not formally recognize postscripts and prefers the use of the <div> tags. Though adding a tag for postscript was considered, we eventually settled on the use of <div> tags to denote this type of text. Because Dickinson’s letters (particularly after the 1850s, but even prior) rarely if ever employ traditional postscripts—such writing appears instead in various places in the margins, at various angles to the main text, and with varying degrees of relationship to the main text—the choice of an extralinear <div> over a <postscript> seems more descriptive and less restrictive.


Upon clicking on a document from the Table of Contents or Search page, viewers encounter a screen that features a page-by-page image of the manuscript, with a transcription on the right and a menu for accessing data on the left. From this screen, readers have several options: to retrieve markup and notes, to view only the image or only the transcription, and/or to click on a link to a floating window with the enlarged image. Readers can also click on the link on the left to view all images in a constellation together on one screen. These digital images of the manuscripts were produced by the repositories in which the documents are housed and are made available in this edition for optimal viewing and web delivery. In a future update, digital images of related print texts will also be added.

These digital surrogates accompany encoded transcriptions—our translations—in order to offer readers representations of the graphical features of Dickinson’s manuscript pages. Important for the reader of this edition to know and keep in mind is that the images are page-by-page, which will often, but not always, fully simulate the experience of reading the manuscripts themselves. When writing is on a single page or leaf, then our simulation very closely maps to reading the manuscript. However, when the writing is across a sheet that is intact, and that has been folded to make two leaves and four pages, our focus on the page does not as closely simulate the reading of the actual physical document. The reader of a manuscript would read the first page (first recto), then open the leaves of the sheet to the second and third pages (first verso and second recto), and then close the opened sheet to read the fourth page (second verso). When the images appear page by page, this handling of the physical document is more difficult to imagine. Readers should consult the version of “Your Riches, taught / me, poverty –” and click on “Alternatively Photographed Image” to see how the second and third pages of a folded sheet face one another. All of the images were processed (i.e., scanned, color-corrected by color-bar, and resized for the screen) in a uniform manner, and are not considered to be “archival-quality” images, which would seriously impede page-load time. As images are not scaled to size, readers are referred to the notes for manuscript size, as well as for the writing’s placement on a page of a leaf or sheet or on a scrap.


Most Dickinson documents are untitled, and we do not wish to impose titles on her manuscripts. However, for the purposes of file description and reference, we display three types of descriptive information, or “titles,” to identify unique documents. First, we create a textual description derived from the initial line or lines of the text. For instance, in one constellation of three different versions of a poem, we use the following descriptors derived from their unique first lines: “A narrow Fellow in / the Grass”; “Dear Friend / Whom / My Dog understood”; and “My Sue / Loo and / Fanny will come / tonight”. We also record a bibliographic reference to the manuscript; in this case, “A 88–13/14 : An Electronic Version,” “BPL 23: An Electronic Version,” and “H B193: An Electronic Version.” Finally, we provide a textual description of the genre, author, recipient, and date or date range; in this instance, “Verse written by Emily Dickinson, early to mid 1860s”; “Letter with Enclosed Verse from Emily Dickinson to Thomas Wentworth Higginson, early March 1866”; “Letter with Embedded Verse from Emily Dickinson to Susan Huntington Dickinson, early 1870s.” In the rare instances that Dickinson provides a title to her manuscript, the title is encoded as the <head> of the text.

“Saying itself in new inflection -” (H 76, FP 333): Transcription

The manuscript itself, when extant, is the source for the document transcription and notes; the source is not the image of the manuscript, which is provided for the user. In some instances, the manuscript is missing or has been destroyed or lost (see, for example, Mabel Loomis Todd’s transcription of a lost version of “What mystery pervades a well!”). In these instances, the electronic text will, first, derive from a transcription by another hand if one exists or from the most authoritative printed source, depending on the editor’s judgment regarding the version that is most closely related to the missing one. If the transcription is not derived from the manuscript itself, the source of the transcription will be recorded in the document header and displayed for the reader.

Primary attention in the displayed transcription is paid to the manuscript as physical object; thus, in the display the physical logic of the text is privileged over its lexical or conceptual logic, though both lexical and bibliographical data are tagged. Our encoded transcription follows the manuscript precisely in regard to line breaks, stanza divisions, capitalization, spelling, punctuation (as much as is possible), and end-of-line hyphenation, and includes Dickinson’s extralinear variants (tagged as <add> ) and marginal writing (tagged as separate <div> s), as well as text added or deleted by Dickinson (see Vetter and McDonald regarding our decisions about representing variants; in Radical Scatters, Werner likewise uses <add> for most variants). The transcription also records handwriting by others, and a description of these additions to the manuscripts is recorded in notes. Document damage and illegible handwriting are also tagged. Any regularization or correction is performed within tags only in order to facilitate better searching and sorting for the user; this means that what appears in the screen display of the transcription is what appears on the manuscript, but that an invisible layer of correction and regularization is sometimes present beneath the surface in order that a search engine can find, for instance, a word broken by end-of-line hyphenation (so that a user can search for “possible” even if “possi-” appears on one line and “ble” on the next).

Additionally, all mandatory and most recommended elements of the TEI are tagged (e.g., including names, dates, places, foreign words and phrases, quotations, and emphasized words and phrases). The tagging of names, dates, places, and foreign words and phrases is invisible to the viewer but aids in searching. Words emphasized in the manuscript are replicated in the transcription to the extent that it is possible to do so on a computer screen; thus, underlining is rendered as underlining (and not as italics, as in most print editions), but text with double underlining, which can’t be easily replicated on a computer screen, is underlined with a reference in the notes to the second underline. Quotation marks, single and double, are rendered as they appear in the manuscript, and are not supplied if Dickinson has not paired them properly. In cases where Dickinson has circled text (see, for instance, “The inundation of / the Spring”), we have noted this, invisibly, in tags so that if this type of rendering becomes possible in the future the transcriptions will show this; in the interim, we have used marks from the ASCII keyboard to approximate this as well as we can.

Because images of the manuscripts will accompany encoded texts whenever possible, users are referred to the image itself to view some physical features that are difficult or impossible for a computer to display accurately (e.g., irregular “dashes,” proportions of vertical and horizontal spacing between letters and words, lines and stanzas, and so forth). Though coming to this decision was difficult, the majority of us agreed that, theoretically speaking, transcription can never capture the material page. Thus, a Dickinson “dash” is recorded as a single hyphen ASCII character. Vertical spacing is represented in units of what amounts to a single hard return. Horizontal spacing between letters and words is not proportionately rendered; when text is indented, the indentations are represented in units of what amounts to single tabs. Likewise, lines drawn between poems or stanzas, typically in fascicles, are merely approximated in their placement; the editors disagree about this practice but in the end determined that it was important to attempt to capture in some way the uniqueness of those lines without rigid standardization. In this initial phase of encoding, these features, idiosyncratic features of Dickinson’s writing, if the editors deem them relevant, are described in notes but not encoded textually; however, the architecture underlying the edition is flexible enough to allow encoding of these features in subsequent stages should editors decide to do so in the future.


Notes, header, and text markup address bibliographic information about the source, including a physical description of the manuscript; references to other relevant manuscript and printed versions, including previous editions; editorial and encoding principles; approximate dating; assignation of genre; and other additional commentary and contextual information. Additionally, as in transcriptions, all mandatory and most recommended elements of TEI are tagged in the notes as well (e.g., including names, dates, places, foreign words and phrases, quotations, and emphasized words and phrases).

What readers will see are several different categories of notes: “Manuscript Description,” which describes the physical manuscript itself; its “History”; “Previous Printings” of the document; “Digital Edition,” which includes information about the status and creation of the electronic edition; “Additional Editorial Notes”; other versions of the writing (or “Constellation Notes,” if applicable); and information regarding how to cite the document (“Cite As”). The XML version of the document is also available. The manuscript description includes not only the location of the manuscript (if it is extant) but as much data as possible about the condition of the document; its size; a description of medium, paper, embossments and watermarks; other handwriting that appears on the pages; and information about enclosures or accompanying material. The other versions and previous printings listed will eventually be linked to pages containing these documents as the edition grows.

“[B]roken mathematics”: Encoding Architecture

The TEI has an inherent and acknowledged bias toward encoding logical and coherent structure, and this was perhaps the chief challenge we faced in encoding Dickinson’s work. That challenge, unfortunately, cannot entirely be met; however, we modified the TEI in order to attend more fully to—at least as much as is possible—the physical features of the manuscript, while still allowing for the deep searching and sorting capabilities facilitated in encoding logical structural elements. Many of these modifications were achieved through adapting the MASTER project’s manuscript description tagset to our particular needs; the MASTER tagset—created by a team led by one of the TEI’s founding members, Lou Burnard—is intended to tag finding aids only, so we have integrated some aspects of it for encoding metadata (instead of whole documents, as the MASTER tagset was originally conceived by its creators).

The set of tags we use in encoding documents is called a Document Type Definition (DTD), and it is based in, and conformant with, the guidelines put forth by the TEI; it is comprised of a mixed base of prose and verse, with tagsets added for linking, figures, analysis, certainty, transcription, critical apparatus, names and dates, and corpora. In addition to the inclusion of several elements from the MASTER project (as well as an <ml/> tag described above), several TEI attribute lists were expanded as well, though some have yet to be used and are present to allow for flexibility in future editions:

The idea behind the construction of this DTD was to make it flexible enough to handle more intricate levels of encoding than we are performing in this initial edition. At some later point, we or other Dickinson scholars might wish to use this edition as a base for their own—one that creates a taxonomies of Dickinson’s “dashes,” for instance, or one that attempts generic classification at a more precise level (the recent debate between Espen Ore and Mats Dahlström on the use of electronic scholarly editions as building blocks for future editions proves especially instructive here).

Editorial Process

Though our original work plan had been to edit the material correspondent-by-correspondent, issuing the Correspondences serially rather than waiting until all of the documents have been edited, the organization of this “digital inquiry” is based on a selection of thirty documents sent to Susan Huntington Dickinson, all of which we have marked up for our originally conceived edition of the complete correspondence (though all, of course, are not featured here). These thirty texts were all printed before 1923 (with the exception of two that were first printed in 1924). Besides the version sent to Susan Dickinson, any other versions of the texts of the edition are also featured. Our work on the edition has been a shared endeavor, bringing together the work of editors with particular expertise in various portions of the Dickinson correspondence, specialists in editing in electronic environments, and graduate and undergraduate students of humanities and of computing. Such a highly collaborative process affords both opportunities as well challenges. In keeping with our desire to open up the editorial process, we are committed to recording editorial disagreements when they cannot be resolved. Disagreements and uncertainties are described in notes, and the TEI permits us to encode these in a combination of <certainty/> and <respons/> tags.

Similarly, we are committed to a policy of maintaining an institutional memory of the project and each part of it. Each document is touched by several virtual hands: those of the editors, the encoders, the proofreaders, the editors at the press. All changes made to the document are recorded in a description of revisions ( <revisionDesc> ), and discoveries and observations made by those other than the guest editors but incorporated into the final document are recorded in the <respons/> tag as well. One of the opportunities afforded editors in an electronic environment is the relative ease of perpetual editing, revisions, and updates. As McGann notes, “It will evolve and change over time, it will gather new bodies of material, its organizational substructures will get modified, perhaps quite drastically” (“Rationale”). Editions—even electronic ones—are created by humans, and though we strive for accuracy, human error is always and inevitably a factor when translating one version of a text to another to yet another. An electronic environment allows for immediate correction and revision, as well as frequent supplements and additions if new materials, discoveries, or observations arise. And using the revision description mechanism to record changes to documents allows for an internal history to be kept of all changes. Werner’s critical observation in Radical Scatters is crucial to keep in mind: “In the future, some of the documents included in the present archive will slip outside of it, while others, not yet marked as fragments, may enter it, producing not new collections but, rather, unforeseen and anomalous orders.”

“What Duplicate - Exist - / What Parallel can be -” (H 18, FP 653): Intentionality and Representation

Finally, it is important to state that by reproducing images of manuscripts, and by transcribing from manuscript, we do not claim to recover Dickinson’s originary or final intentions. The manuscript itself cannot participate in such a claim. Texts—original and edited, in print and in digital form—are products of particular socio-historical moments; as Walter Benjamin has observed, “Even the most perfect reproduction of a work of art is lacking in one element: its presence in time and space, its unique existence at the place where it happens to be” (220). Furthermore, we do not believe that a “true” diplomatic transcription of Dickinson documents for electronic transmission is an achievable goal. Transcription is always translation, and our transcriptions are our best approximations; we believe that no transcription or image, electronic or otherwise, can ever capture visual or graphical data with sufficient accuracy to represent the original. In fact, as editors we amiably disagree about the conclusions to be drawn from our work even as we remain unanimously committed to the value of manuscript scholarship for the study of Dickinson and nineteenth-century literature and culture.

III. A Brief Note on the Principles of Selection for the Thirty Texts Featured Here

Harvard University Press asked that we select thirty texts for a limited digital edition rather than publish an edition of all of the writings Dickinson “published” herself by distributing to various correspondents. Since we consider the writings to Susan Huntington Dickinson the primary correspondence of Emily Dickinson—besides being three times the size of the next most frequently addressed correspondent (Higginson), the corpus to Susan contains the only instance of Dickinson changing a poem at another’s behest—we chose thirty texts from this main body of Dickinson’s epistolary writing to serve as the foundation for this edition. Also, had we chosen writings from Dickinson’s manuscript books or from the loose sheets found after her death or those sent to another correspondent as our base, the diversity of manuscripts featured in this edition would have been much more limited. The correspondence with Susan Dickinson ranges across decades. Unlike the materials found in the manuscript books, on the loose sheets, and in other correspondences, the writings to Susan Dickinson are on different types and sizes of paper, some formal, some scraps (such as “The face we choose to miss,” which is on irregular paper with Susan’s transcription on the back). Some writings to Susan are “finished” and others are in draft form such as “Frequently the woods are pink” and “Who were the Father and the Son”. Some of the writings sent to Susan were intentionally mutilated. Some feature drawings or cut-outs. And the line between verse and prose in the correspondence with Susan often blurs. Thus, by working from writings to Susan Dickinson in order to make this radiant edition, we were able to feature Dickinson’s manuscript writings in all of their known diversities.

Undoubtedly, placing the material text at the center of our editorial theory and practice has implications for the study of Dickinson, and we are excited by possible new avenues of exploration for scholars and readers, teachers and students, who previously have not had access to good facsimile images of Dickinson’s writing. However, these new avenues do not constitute the only pathways to research, teaching, and reading. We would like to see this edition read in tandem with the excellent printed scholarly editions available on library shelves, and we welcome user input, since one of our primary goals is to be in dynamic dialogue with our readers.

Join us on Facebook

Works Cited and Consulted

Benjamin, Walter. Illuminations. Ed. Hannah Arendt. Trans. Harry Zohn. New York: Schocken Books, 1968.

Bianchi, Martha Dickinson. The Life and Letters of Emily Dickinson. Boston and New York: Houghton Mifflin, 1924.

Bryant, John. The Fluid Text: A Theory of Revision and Editing for Book and Screen. Ann Arbor: University of Michigan Press, 2002.

__________. “Witness and Access: The Uses of the Fluid Text.” Textual Cultures: Texts, Contexts, Interpretation 2.1 (Spring 2007): 16–42.

Dahlström, Mats. “How Reproductive is a Scholarly Edition?” Literary and Linguistic Computing 19.1 (2004): 17–33.

Dickinson, Emily Elizabeth. Manuscripts at the Amherst College Library are cited using “A” and the library catalog number. Manuscripts at the Boston Public Library are cited using “BPL” and the library catalog number. Manuscripts at the Houghton Library are cited using “H” and the library catalog number.

Folsom, Ed. “Database as Genre: The Epic Transformation of the Archives.” PMLA 122.5 (October 2007): 1571–1579.

Franklin, R. W., ed. The Manuscript Books of Emily Dickinson. Cambridge, Mass., and London: Belknap Press of Harvard University Press, 1981.

__________. The Poems of Emily Dickinson: Variorum Edition. Cambridge, Mass., and London: Belknap Press of Harvard University Press, 1998. References to poems in this text are indicated by “FP” and the number assigned by Franklin.

Grigely, Joseph. Textualterity: Art, Theory, and Textual Criticism. Ann Arbor: University of Michigan Press, 1995.

Hart, Ellen Louise, and Martha Nell Smith, eds. Open Me Carefully: Emily Dickinson’s Intimate Letters to Susan Huntington Dickinson. Ashfield, Mass.: Paris Press, 1998. References to writings in this text are indicated by “OMC” and the number assigned by Hart and Smith.

Jackson, Virginia. Dickinson’s Misery: A Theory of Lyric Reading. Princeton, N.J. and Oxford: Princeton University Press, 2005.

Johnson, Thomas H., ed. The Poems of Emily Dickinson. Cambridge, Mass.: Belknap Press of Harvard University Press, 1955. References to poems in this text are indicated by “JP” and the number assigned by Johnson.

__________ and Theodora Ward, eds. The Letters of Emily Dickinson. Cambridge, Mass.: Belknap Press of Harvard University Press, 1958. References to poems in this text are indicated by “JL” and the number assigned by Johnson and Ward.

Latour, Bruno, and Steve Woolgar. Laboratory Life: The Construction of Scientific Facts. Princeton, N.J.: Princeton University Press, 1979. Second ed. 1986.

McGann, Jerome. “Database, Interface, and Archival Fever” (Response to Ed Folsom’s “Database as Genre”). PMLA 122.5 (October 2007): 1588–1592.

__________. Radiant Textuality: Literature after the World Wide Web. New York: Palgrave, 2001.

__________. “The Rationale for HyperText.” 29 March 2005 http://www.iath.virginia.edu/public/jjm2f/rationale.html.

__________. The Textual Condition. Princeton, N.J.: Princeton University Press, 1991.

McGill, Meredith L. “Remediating Whitman” (Response to Ed Folsom’s “Database as Genre”). PMLA 122.5 (October 2007): 1592–1596.

Ore, Espen S. “Monkey Business—or What Is an Edition?” Literary and Linguistic Computing 19.1 (2004): 35–44.

Smith, Martha Nell. ‘Because the Plunge from the Front Overturned Us’: The Dickinson Electronic Archives Project.” Studies in the Literary Imagination 32.1 (1999): 1–19.

__________. “Computing: What’s American Literary Study Got to Do with IT?” American Literature 74.4 (December 2002): 833–857.

__________. “Dickinson’s Manuscripts.” The Emily Dickinson Handbook, ed. Gudrun Grabher, Roland Hagenbüchle, Cristanne Miller. 113–137. Amherst, Mass., and Boston: University of Massachusetts Press, 1998; pbk edition 2004.

__________. Rowing in Eden: Rereading Emily Dickinson. Austin: University of Texas Press, 1992.

__________. “The Human Touch, Software of the Highest Order: Revisiting Editing as Interpretation.” Textual Cultures 2.1 (Spring 2007): 1–15.

__________ and Mary Loeffelholz, eds. Companion to Emily Dickinson. Oxford, Malden, Mass., and Carlton, Victoria, Australia: Blackwell Publishing, 2008.

Socarides, Alexandra. “The Poetics of Interruption: Dickinson, Death, and the Fascicles.” A Companion to Emily Dickinson, ed. Martha Nell Smith and Mary Loeffelholz. 309–333. Malden, Mass., Oxford, England, and Victoria, Australia: Blackwell Publishing, 2008.

Sperberg-McQueen, C. M. “Text in the Electronic Age: Textual Study and Text Encoding, with Examples from Medieval Texts.” Literary and Linguistic Computing 6.1 (1991): 34–46.

__________, and L. Burnard, eds. TEI P4: Guidelines for Electronic Text Encoding and Interchange. 2002. Text Encoding Initiative Consortium. XML Version: Oxford, England, Providence, R.I. Charlottesville, Va., Bergen, Norway. 29 March 2005 http://www.tei-c.org/P4X/.

Stallybrass, Peter. “Against Thinking” (Response to Ed Folsom’s Database as Genre). PMLA 122.5 (October 2007): 1580–1587.

Vetter, Lara. “Editing Dickinson in an Electronic Environment.” A Companion to Emily Dickinson, ed. Martha Nell Smith and Mary Loeffelholz. 437–452. Malden, Mass., Oxford, England, and Victoria, Australia: Blackwell Publishing, 2008.

__________, and Jarom McDonald. “Witnessing Dickinson’s Witnesses.” Literary and Linguistic Computing 18.2 (2003): 151–65.

Werner, Marta, ed. Radical Scatters: Emily Dickinson’s Fragments and Related Texts, 1870–1886. Ann Arbor: University of Michigan Press, 1999; republished June 2007, Lincoln: Center for Digital Research in the Humanities, University of Nebraska.